BoxComm - Commentary Rhythm (t-IoU): leaderboard

Metric: Mean temporal IoU (0-1) between the predicted and the human-annotated speech intervals of each commentary category, averaged over the three categories, in streaming commentary rhythm assessment on BoxComm-Eval: the model watches each match in short chunks and comments freely with a generic commentator prompt, GPT-5.2 classifies each generated sentence as play-by-play, tactical or contextual, and the timing is compared with the human broadcast commentary; zero-shot; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 2 models tracked.

Top models

Interactive version: theaggregate.ai/benchmark?slug=boxcomm-commentary-rhythm-t-iou · How It Works · Data refreshed daily, snapshot 2026-10-07.